Goldman Sachs: High Yields on Korea's 20-Year Bonds May Be a Value Trap
Nashnova编辑部
Goldman Sachs warns that the high yield on Korea's 20-year government bonds may be a value trap — the structural demand shortfall driving long-end weakness remains unchanged, and a high yield alone is not a reason to buy.
Long-end yields look cheap — why does Goldman say hold off?
Korea's 30-year government bond yield is already about 20 to 25 basis points cheaper than fair value, which looks attractive on the surface.
Goldman calls this a "value trap" — the cheapness is structural, not a temporary mispricing.
This means → cheap does not equal worth buying. If the forces depressing prices persist, "cheap" just keeps getting cheaper.
What is driving the structural weakness?
The core issue is missing demand: long-dated bonds cannot find enough buyers.
First, equities and other risk assets offer higher nominal returns, leaving investors little incentive to lock in 30-year bonds.
Second, global investors demand higher real returns. Goldman stresses that roughly 80% of the rise in U.S. long-end yields comes from real rates, not inflation expectations — and this pressure has spilled into Korea.
In plain terms = Korea's bonds are not broken on their own. The global environment of "nobody wants long-duration debt" is doing the damage.
Can the Finance Ministry's countermeasures help?
The ministry has adjusted its issuance schedule, focusing policy attention on the 30-year tenor to ease primary-market supply pressure.
Goldman's verdict: these are marginal fixes and do especially little for the 20-year point on the curve.
This means → the supply side is already doing what it can, but the problem sits on the demand side — supply-side tweaks cannot close a structural demand gap.
Is the 20-year "cheapness" real, or a pricing distortion?
Goldman says the real problem lies on the interest-rate swap (IRS) curve — a benchmark derived from contracts where institutions exchange fixed for floating rates — where the 20-year point is mispriced.
On the nominal government-bond spline curve alone, the 20-year tenor does not stand out as unusually cheap.
In plain terms = the 20-year bond looks cheap mainly because the IRS curve distortion "projects" cheapness onto it, not because the bond itself is clearly mispriced.
A relative-value trade betting on 20-year normalization had formed earlier, but that spread has largely converged.
What just happened in the market — can the move continue?
Concentrated selling hit Korea's long end, compounded by offshore funds paying fixed on long-tenor IRS, steepening the 10-to-20-year curve by about 4 basis points.
The rise in 20-year rates pushed the 10-year-10-year forward swap rate — the market-implied rate from year 10 to year 20 — up roughly 30 basis points in about 10 days, from 3.70% to 4.00%.
Goldman cautions: a move that fast triggers hedging from volatility desks, creating offsetting supply that caps how fast long-end rates can keep rising.
This means → a short-term technical pullback is possible, but medium-term risk still tilts toward higher rates.
Why don't institutional investors buy the dip at these yields?
Korean asset managers and insurers sense some "value" even at 4.50% yields, yet cannot meaningfully increase 30-year exposure.
The constraints stack four deep: excessive duration risk, high capital charges, liability-side mismatch, and insufficient risk budget.
Goldman notes that Korea's persistent long-bond weakness increasingly resembles Japan's pattern.
In plain terms = like Japan's long-dated bonds, the issue is not that nobody sees the value — it is that the institutions able to buy are structurally constrained from acting. Until that demand-side gap is fixed, high yields remain a case of "look but don't touch."
Content is for reference only, not financial advice.